Systems and methods for soil clod detection

1. A method for determining soil clods within a field, the method comprising:

receiving, with a computing system, one or more images depicting an imaged portion of an agricultural field, the imaged portion of the field being at least partially represented by a plurality of pixels within the one or more images;

classifying, with the computing system, a portion of the plurality of pixels that are associated with soil within the imaged portion of the field as soil pixels, each soil pixel being associated with a respective pixel height;

identifying, with the computing system, a local maximum as a location of a candidate soil clod based on the pixel heights of the soil pixels;

generating, with the computing system, a first ray extending from the local maximum in a first direction until opposing endpoints are determined based on a detected edge condition;

generating, with the computing system, a second ray from the local maximum in a second direction that is perpendicular to the first ray until opposing endpoints are determined based on a detected edge condition; and

determining, with the computing system, whether a soil clod is present within the imaged portion of the field based on a soil clod ratio of a length of the first ray to a length of the second ray or one or more pixel heights along the first ray or the second ray of the candidate soil clod.

2. The method of claim 1 , wherein the one or more images comprise a depth image of the imaged portion of the agricultural field, each of the plurality of pixels within the depth image being associated with a respective pixel height.

3. The method of claim 2 , further comprising:

generating, with the computing system, a slope-corrected depth image in which the depth image is corrected based on an estimated slope of the imaged portion of the field, the pixel height of each of the plurality of pixels deriving from the slope-corrected depth image.

4. The method of claim 2 , wherein the detected edge condition is a location in which the first ray or the second ray reaches a void in the depth image.

5. The method of claim 2 , wherein classifying the portion of the plurality of pixels that are associated with soil within the imaged portion of the field as soil pixels, identifying a local maximum as a location of a candidate soil clod based on the pixel heights of the soil pixels, generating the first ray extending from the local maximum, generating the second ray from the local maximum in a second direction that is perpendicular to the first ray, and determining whether the soil clod is present within the imaged portion of the field are completed for a plurality of copies of the depth image, with each of the plurality of copies of the depth image downscaled a different amount, and wherein each clod calculated across each of the plurality of copies of the depth image are combined into a single list, ordered by size.

6. The method of claim 1 , wherein the detected edge condition is a location in which an inflection point is detected based on the pixel height for each respective pixel within the soil pixels.

7. The method of claim 1 , wherein the detected edge condition is a location in which a non-soil pixel is reached.

8. The method of claim 1 , wherein determining whether a soil clod is present within the imaged portion of the field based at least in part on a soil clod ratio of the length of the first ray to the second ray further comprises determining whether the soil clod ratio exceeds a soil clod ratio threshold.

9. The method of claim 1 , further comprising: placing each of the soil clods in two or more bins based on an area of each soil clod.

10. The method of claim 1 , further comprising:

generating a soil clod map of each of the soil clods within the imaged portion, wherein each of the soil clods is positioned within the soil clod map in a position that correlates to its position in the imaged portion of the field.

11. The method of claim 1 , wherein receiving the one or more images comprises receiving the one or more images from an imaging device supported relative to an agricultural machine, the one or more images being captured by the imaging device as the agricultural machine travels across the field.

12. The method of claim 11 , further comprising initiating, with the computing system, a control action associated with adjusting an operation of the agricultural machine based at least in part on the determination that a quantity of soil clods or a size of soil clods within the imaged portion of the field exceeds a predefined threshold.

13. A system for determining one or more soil clods within a field, the system comprising:

an imaging device configured to capture one or more images depicting an imaged portion of the field, the imaged portion of the field being at least partially represented by a plurality of pixels within the one or more images; and

a computing system communicatively coupled to the imaging device, the computing system including a processor and associated memory, the memory storing instructions that, when implemented by the processor, configure the computing system to:

receive the one or more images from the imaging devices;

classify a portion of the plurality of pixels that are associated with soil within the imaged portion of the field as soil pixels, each soil pixel being associated with a respective pixel height;

determine one or more local maximum based on the pixel heights of the soil pixels;

identify each of the one or more local maximum as a location of a candidate soil clod;

generate a first ray and a second ray extending in perpendicular directions from one another, with at least one of the first ray or the second ray intersecting the local maximum; and

determining whether the soil clod is present within the imaged portion of the field based on a soil clod ratio of a length of the first ray to a length of the second ray or one or more pixel heights along the first ray or the second ray of the candidate soil clod.

14. The system of claim 13 , wherein the one or more images comprise a depth image of the imaged portion of the field, with each of the plurality of pixels within the depth image being associated with a respective pixel height.

15. The system of claim 14 , wherein the computing system is further configured to determine an area of each soil clod based on the length of the first ray and the length of the second ray and determine a quantity of soil clods within the captured image having an area that is equal to or greater than a predefined threshold.

16. The system of claim 15 , wherein the computing system is further configured to generate a map of each soil clod within the imaged portion based on the local maximum of each soil clod and the generated boundary of each soil clod.

17. A method for determining soil clods within a field, the method comprising:

receiving, with a computing system, one or more images depicting a first imaged portion of an agricultural field, the first imaged portion of the field being at least partially represented by a plurality of pixels within the one or more images;

classifying, with the computing system, a portion of the plurality of pixels that are associated with soil within the first imaged portion of the field as soil pixels, each soil pixel being associated with a respective pixel height;

generating a first ray and a second ray extending in perpendicular directions from one another from each local maximum within the soil pixels of the first imaged portion, with at least one of the first ray or the second ray intersecting the local maximum; and

determining, with the computing system, whether a soil clod is present within the imaged portion of the field based at least in part on a soil clod ratio of a length of the first ray to a length of the second ray.

18. The method of claim 17 , further comprising:

determining a quantity of soil clods in the first imaged portion of the field to a quantity of soil clods in a second imaged portion of the field.

19. The method of claim 17 , wherein receiving the one or more images comprises receiving the one or more images from an imaging device supported relative to an agricultural machine, the one or more images being captured by the imaging device as the agricultural machine travels across the field, the method further comprising:

initiating a control action associated with adjusting an operation of the agricultural machine when the soil clod ratio exceeds a predefined threshold.